| Accurate in vivo 3-D joint kinematics are necessary to precisely diagnose, assess and treat musculoskeletal injuries, impairments and pathologies. Nearly one in seven Americans [Praemer et al., 1999] are affected by physical injuries which often limit an individual's ability to perform routine tasks, decreases productivity and may affect overall quality of life. The cost of treating these injuries, impairments and pathologies and the burden they impose on the health care systems are significant with the number of such conditions continuously increasing.;Cine Phase Contrast (fine PC) Magnetic Resonance Imaging (MRI) is a non-invasive, in vivo technique which is routinely used to determine skeletal kinematics such as those for entire patello-tibio-femoral system [Rebmann and Sheehan, 2003]. In such studies, the attitudes of the patellofemoral and tibiofemoral joints were individually quantified using the position trajectories of three or more points on each bone derived through the integration of the fast-PC data. This integration was based on Fourier techniques, which are limited by assumptions in regards to out-of-plane motion and the minimal number of points used in the analysis.;The purpose of this dissertation was to develop a new integration algorithm, the Precise Motion Determination Algorithm (PMD) that improves integration accuracy by removing the primary limitations of the Fourier integration techniques. The heart of the PMD was a least squares routine for determining the six transformation parameters that describe the change in orientation and position a rigid body undergoes between any two epochs in a cycle.;The PMD and the original Fourier integration technique were tested using cine PC MRI phantom data, enabling the performance of both algorithms to be compared to a gold standard. In addition, numerous synthetic datasets were developed and tested with both algorithms in order to determine how well both performed under different levels and types (i.e., Gaussian, Laplacian and Uniform) of noise. The PMD's performance characteristics for estimating the transformation parameters under noisy conditions and improved accuracy of tracking multiple in and out-of-plane points simultaneously, show that using the PMD with cine PC MRI data is an improved and promising technique for quantifying 3-D joint kinematics. |
